SIMULATION STUDY OF MUON ACCELERATION USING RFQ FOR ANEW MUON G-2 EXPERIMENT AT J-PARC

Y. Kondo∗, K. Hasegawa, JAEA, Tokai, Naka, Ibaraki, 319-1195, JapanM. Otani, T. Mibe, N. Saito, KEK, Oho, Tsukuba, 305-0801, Japan

R. Kitamura, U. Tokyo, Tokyo, 113-0033, Japan

AbstractA new muon g-2 experiment is planned at J-PARC. In

this experiment, ultra cold muons will be generated and ac-celerated using a linear accelerator. As the irst acceleratingstructure, an RFQ will be used. We are planning to use aspare RFQ of the J-PARC linac for the irst acceleration test.In this paper, simulation studies of this muon accelerationtest are presented. A design study of a muon dedicated RFQis also shown.

INTRODUCTIONThe J-PARC E34 experiment aims to investigate parti-

cle physics beyond the standard model by measuring themuon anomalous magnetic moment (g-2) and the electricdipole moment (EDM) with precision of 0.1 ppm and 1 ×10−21�⋅cm, respectively [1]. E34 will be conducted at the H-line [2] of J-PARC muon science facility. This experimentneeds a low-emittance muon beam, and to this end, the ideaof reacceleration of cooled muon is utilized. The reacceler-ation of cooled muon is already proposed in [3], in whicha wedge-shaped energy absorber is used to cool the muons.On the contrary, we are planning to use ultra cold muons(UCMs) generated by laser-ionization of thermal muoniums(Mu: �+e−) form an silica-aerogel target [4]. GeneratedUCMs are accelerated using a muon linac. As the irst step,a spare RFQ of J-PARC linac (RFQ II [5]) will be used as afront-end accelerator. The resonant frequency of RFQ II is324 MHz.

Prior to construct the muon linac, we are planning to con-duct a muon acceleration experiment using RFQ II at theH-line. The accelerated particle for E34 experiment is �+,however, the ultra cold muon source will not be availablewhen the initial stage acceleration experiment will be per-formed. Therefore, we are developing a negative muonium(Mu−) source [6], with witch the Mu−’s are generated bydecelerating the muons from the H-line using an aluminumdegrader.

RFQ II is originally designed to accelerate H−’s, and themass of the H−is nine times larger than that of muon. To ac-celerate muons using RFQ II, the power should be reducedto 1/80 of the design power; this is very ineicient. There-fore, we are planning to develop a muon dedicated RFQ(muRFQ), and replace RFQ II when the following acceler-ators are ready.

In this paper, simulation studies of muon acceleration us-ing RFQs are described.

∗ yasuhiro.kondo@j-parc.jp

UCM ACCELERATION USING J-PARCRFQ II

For the simulation of J-PARC RFQ II, PARMTEQM [7]is used because it was used for the optics design of RFQ II.To accelerate muons, the original input ile should be modi-ied as follows; 1) The particle mass is deined as muon. 2)Input and output energy are converted from the �s of theH−. The muon mass and the input and output energy aresummarized in Table 1 together with the original parame-ters for H− acceleration.

Table 1: Parameter Conversion from H to− �Beam species H− �Mass (MeV/c2) 939.302 105.658Injection � 0.010318Injection energy (keV) 50.000 5.625Extraction � 0.079732Extraction energy (MeV) 3.000 0.337Inter vane voltage (V) 82.879 9.324Nominal power (kW) 330 4.177

In this simulation, a particle distribution simulated usingGEANT4 [6] is used, as shown in Fig. 1. The phase distri-bution is assumed to be uniform.

x (mm)

xp (

mra

d)

y (mm)

yp (

mra

d)

energy (MeV)

counts

∆φ (deg)

counts

Figure 1: UCM distribution at the RFQ II entrance. Theellipses in upper igures represent the matched ellipses of��,�,��� = 0.167� mm mrad.

6th International Particle Accelerator Conference IPAC2015, Richmond, VA, USA JACoW PublishingISBN: 978-3-95450-168-7 doi:10.18429/JACoW-IPAC2015-THPF045

4: Hadron AcceleratorsA09 - Muon Accelerators and Neutrino Factories

THPF0453801

Cont

entf

rom

this

wor

km

aybe

used

unde

rthe

term

soft

heCC

BY3.

0lic

ence

(©20

15).

Any

distr

ibut

ion

ofth

isw

ork

mus

tmai

ntai

nat

tribu

tion

toth

eau

thor

(s),

title

ofth

ew

ork,

publ

isher

,and

DO

I.

Figure 2: PARMTEQM simulation result of UCM acceler-ation.

Figure 3: Particle distribution of muons at the RFQ II exit.

Figs 2 and 3 show the simulation results. The horizon-tal and vertical normalized rms emittances at the RFQ exitare 0.29 and 0.17 � mm mrad, respectively, and the en-ergy spread is 3.4 keV. The simulated transmission is 94.5%.The number of the cells of RFQ II is 295, thus the transienttime through the RFQ is 455 ns. Therefore the survival rateof the muon through the RFQ is estimated to be 0.81, andthe total transmission is 77%.

SIMULATION OF MU− ACCELERATIONEXPERIMENT

The simulation for Mu− acceleration experiment hasbeen done also using PARMTEQM. Fig. 4 shows theparticle distribution at the RFQ entrance simulated usingGEANT4. In this experiment, the muons from the H-lineare decerelated using an Al degrader, therefore, the trans-verse momentum of generated Mu− is much larger than thatof UCM case, thus the emittance at the RFQ entrance is verylarge. The phase distribution is also assumed to be uniform.

Fig. 5 represents the phase plots at the exit of the RFQ.The transmission through the RFQ is 9.5%, and the hori-zontal and vertical normalized rms emittances at the RFQexit are 0.40 and 0.44 � mm mrad, respectively. Assumingthe 1 MW proton beam is provided to the muon productiontarget, the number of �+ on the Mu− production target is es-timated to be 3×106 /s. A conversion eiciency to Mu− andthe transportation eiciency are assumed to 1.8 × 10−4 [8]and 20%, respectively, therefore, the yield at the RFQ exit

x (mm)

xp (

mra

d)

y (mm)

yp (

mra

d)

energy (MeV)

counts

∆φ (deg)

counts

Figure 4: Input distribution of the Mu− simulation.

Figure 5: Phase plots of Mu− at the RFQ exit.

will be 8 Hz. To detect the accelerated Mu− will be feasiblewith this rate.

MUON DEDICATED RFQIn this section, preliminary design of muRFQ is shown.

We used RFQGEN [9] for the beam dynamics design andthe particle simulation of muRFQ. Fig. 6 shows the de-signed cell parameters of muRFQ.

To increase the acceleration eiciency, the inter-vanevoltage is increased compared to that of RFQ II. Increasedvoltage makes the focusing force too strong, thus, the boreradius is also increased to reduce the focusing strength. Inthis design, the injection energy of muon is 30 keV and theextraction energy is 0.34 MeV.

Fig. 7 shows the evolution of the distribution trough theRFQ, and Fig. 8 represents the phase plots at the RFQ exit.The input transverse distribution for this simulation is 0.17� mm mrad (normalized, rms) water-bag. As for the lon-gitudinal distribution, no energy spread and uniform phasedistribution are assumed.

In Table 2, the design parameters and the simulation re-sults of muRFQ are summarized together with those of

6th International Particle Accelerator Conference IPAC2015, Richmond, VA, USA JACoW PublishingISBN: 978-3-95450-168-7 doi:10.18429/JACoW-IPAC2015-THPF045

THPF0453802

Cont

entf

rom

this

wor

km

aybe

used

unde

rthe

term

soft

heCC

BY3.

0lic

ence

(©20

15).

Any

distr

ibut

ion

ofth

isw

ork

mus

tmai

ntai

nat

tribu

tion

toth

eau

thor

(s),

title

ofth

ew

ork,

publ

isher

,and

DO

I.

4: Hadron AcceleratorsA09 - Muon Accelerators and Neutrino Factories

Figure 6: Designed cell parameters of the muRFQ usingRFQGEN.

Figure 7: Evolution of the distribution in muRFQ.

RFQ II. As shown in Table 2, the vane length is 1/3 com-pared to RFQ II. Although the power dissipation is 50 timeslarger than that of RFQ II, it is a great merit because the fab-rication of 1 m RFQ is much easier than that of 3 m RFQ.The number of cells is 72 and it takes 111 ns to pass throughthe RFQ. The transmission through the RFQ is 96%, it is alittle bit lower than that of the RFQ, but the decay rate isimproved from 20% to 5%, then the total eiciency is im-proved from 80% to 90%. The transverse emittance growthratio of the present design is 20%, so it is necessary moreoptimization of the design.

Figure 8: Phase plots of extracted particles from muRFQ.

Table 2: Design Parameters of RFQ II and Muon RFQ

RFQ II muRFQ

Resonant frequency (MHz) 324Injection energy (keV) 5.625 30Extraction energy (MeV) 0.34Inter vane voltage (kV) 9.3 138Power (kW) 4.2 200Number of cells 295 72Length (m) 3.17 1.10Transmission (%) 99.9 96.2Input emittance 0.17(� mm mrad, normalized, rms)Output emittance x 0.17 0.20(� mm mrad, normalized, rms)Output emittance z 0.021 0.25(� MeV deg, rms)

SUMMARYWe performed simulation studies of RFQs for a new

muon g-2 experiment at J-PARC. In the irst phase of theexperiment, J-PARC RFQ II will be used. A particle simula-tion of ultra-cold-muon acceleration using RFQ II has beendone, and the transmission is 94.5%. Prior to the construc-tion of the muon linac, a muonium acceleration test willbe conducted. Simulation of the muonium acceleration isalso done and the estimated muonium rate at the RFQ exitis 8 Hz. Design study of muon dedicated RFQ is going on,and we have demonstrated a compact 1-m muon RFQ de-sign.

REFERENCES[1] “J-PARC E34 Conceptual Design Report” (2011).[2] N. Kawamura et al., “H line; A Beamline for Fundamen-

tal Physics in J-PARC”, Proceedings of USM2013, 010112(2014).

[3] H. Miyadera et al., “Design of muon accelerators for an ad-vanced muon facility”, Proceedings of PAC07, Albuquerque,New Mexico, USA, 3032–3034 (2007).

[4] P. Bakule et al., “Measurement of muonium emission fromsilica aerogel”, Prog. Theor. Exp. Phys, 103C01 (2013).

[5] Y. Kondo et al., “High-power test and thermal characteristicsof a new radio frequency quadrupole cavity for the japan pro-ton accelerator research complex linac”, Phys. Rev. ST Accel.Beams 16, 040102 (2013).

[6] M. Otani et al., “Development of muon linac for the muong-2/EDM experiment at J-PARC”,IPAC15,These Proceedings.

[7] K. R. Crandall, et al., “RFQ design codes”, LA-UR-96-1836Revised December 7, 2005, Los Alamos National Laboratory(1996).

[8] A. Kuang et al., “Formation of the negative muonium ionand charge-exchange processes for positive muons passingthrough thin metal foils”, Phys. Rev. A 39 (12), 6109–6123(1989).

[9] Linear Accelerators, “RFQGEN”.

6th International Particle Accelerator Conference IPAC2015, Richmond, VA, USA JACoW PublishingISBN: 978-3-95450-168-7 doi:10.18429/JACoW-IPAC2015-THPF045

4: Hadron AcceleratorsA09 - Muon Accelerators and Neutrino Factories

THPF0453803

Cont

entf

rom

this

wor

km

aybe

used

unde

rthe

term

soft

heCC

BY3.

0lic

ence

(©20

15).

Any

distr

ibut

ion

ofth

isw

ork

mus

tmai

ntai

nat

tribu

tion

toth

eau

thor

(s),

title

ofth

ew

ork,

publ

isher

,and

DO

I.